Identification, by systematic RNA sequencing, of novel candidate biomarkers and therapeutic targets in human soft tissue tumors

Human sarcomas comprise a heterogeneous group of more than 50 subtypes broadly classified into two groups: bone and soft tissue sarcomas. Such heterogeneity and their relative rarity have made them challenging targets for classification, biomarker identification, and development of improved treatment strategies. In this study, we used RNA sequencing to analyze 35 primary human tissue samples representing 13 different sarcoma subtypes, along with benign schwannoma, and normal bone and muscle tissues. For each sarcoma subtype, we detected unique messenger RNA (mRNA) expression signatures, which we further subjected to bioinformatic functional analysis, upstream regulatory analysis, and microRNA (miRNA) targeting analysis. We found that, for each sarcoma subtype, significantly upregulated genes and their deduced upstream regulators included not only previously implicated known players but also novel candidates not previously reported to be associated with sarcoma. For example, the schwannoma samples were characterized by high expression of not only the known associated proteins GFAP and GAP43 but also the novel player GJB6. Further, when we integrated our expression profiles with miRNA expression data from each sarcoma subtype, we were able to deduce potential key miRNA–gene regulator relationships for each. In the Ewing's sarcoma and fibromatosis samples, two sarcomas where miR-182-5p is significantly downregulated, multiple predicted targets were significantly upregulated, including HMCN1, NKX2-2, SCNN1G, and SOX2. In conclusion, despite the small number of samples per sarcoma subtype, we were able to identify key known players; concurrently, we discovered novel genes that may prove to be important in the molecular classification of sarcomas and in the development of novel treatments.